INDEX

Explanations

names of television shows, characters, and media-related terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lesi

-0.15

atural

-0.15

 odds

-0.14

alla

-0.14

ItemSelected

-0.14

rze

-0.13

 Pitch

-0.13

 cogn

-0.13

egative

-0.13

apult

-0.13

POSITIVE LOGITS

ofday

0.15

ë¹

0.14

usercontent

0.14

uppen

0.14

ÑĮÑİ

0.14

.rl

0.14

timeofday

0.14

Ned

0.14

åª

0.13

lib

0.13

Activations Density 0.094%

names of television shows, characters, and media-related terms

No Comments

No Known Activations